Influence of crystallization state and microstructure on the chemical durability of cerium-neodymium mixed oxides

Article abstract of DOI:10.1021/ic201269c

To underline the potential links between the crystallization state and the microstructure of powdered cerium-neodymium oxides and their chemical durability, several Ce^IV_1-xNd^III _xO_2-x/2 mixed dioxides were prepared in various operating conditions from oxalate precursors and then leached. The powdered samples were first examined through several physicochemical properties (crystallization state and associated crystallite size, reactive surface area, porosity...). The dependence of the normalized dissolution rates on various parameters (including temperature, nitric acid concentration, crystallization state) was examined for pure CeO₂ and Ce_1-xNd_xO_2-x/2 solid solutions (with x = 0.09 and 0.16). For CeO₂, either the partial order related to the proton activity (n = 0.63) or the activation energy (E _A = 37 kJ?mol^-1) suggested that the dissolution was mainly driven by surface reactions occurring at the solid-liquid interface. The chemical durability of the cerium-neodymium oxides was also strongly affected by chemical composition. The initial normalized dissolution rates were also found to slightly depend on the crystallization state of the powders, suggesting the role played by the crystal defects in the dissolution mechanisms. On the contrary, the crystallite size had no important effect on the chemical durability. Finally, the normalized dissolution rates measured near the establishment of saturation conditions were less affected, which may be due to the formation of a gelatinous protective layer at the solid/liquid interface.

Full text of DOI:10.1021/ic201269c

ARTICLE
pubs.acs.org/IC
Influence of Crystallization State and Microstructure on the Chemical
Durability of CeriumꢀNeodymium Mixed Oxides
Laurent Claparede,†,‡ Nicolas Clavier,† Nicolas Dacheux,*,† Philippe Moisy,‡ Renaud Podor,† and
Johann Ravaux†
†ICSM, UMR5257 CNRS/CEA/UM2/ENSCM, Site de Marcoule, B^at. 426, BP 17171, 30207 Bagnols/Cꢀeze, France
‡CEA, Nuclear Energy Division, RadioChemistry & Processes Department, BP 17171, 30207 Bagnols/Cꢀeze, France
ABSTRACT: To underline the potential links between the crystallization state
and the microstructure of powdered ceriumꢀneodymium oxides and their
chemical durability, several CeIV1ꢀxNdIIIxO2ꢀx/2 mixed dioxides were prepared
in various operating conditions from oxalate precursors and then leached. The
powdered samples were first examined through several physicochemical proper-
ties (crystallization state and associated crystallite size, reactive surface area,
porosity...). The dependence of the normalized dissolution rates on various
parameters (including temperature, nitric acid concentration, crystallization
state) was examined for pure CeO2 and Ce1ꢀxNdxO2ꢀx/2 solid solutions (with
x = 0.09 and 0.16). For CeO2, either the partial order related to the proton activity
(n = 0.63) or the activation energy (EA = 37 kJ molꢀ1) suggested that the
3
dissolution was mainly driven by surface reactions occurring at the solidꢀliquid
interface. The chemical durability of the ceriumꢀneodymium oxides was also
strongly affected by chemical composition. The initial normalized dissolution rates were also found to slightly depend on the
crystallization state of the powders, suggesting the role played by the crystal defects in the dissolution mechanisms. On the contrary,
the crystallite size had no important effect on the chemical durability. Finally, the normalized dissolution rates measured near the
establishment of saturation conditions were less affected, which may be due to the formation of a gelatinous protective layer at the
solid/liquid interface.
such physicochemical properties of the mixed dioxides need to be
optimized in order to fit with the severe environments planned in
the Gen IV-type reactor concepts.11
1. INTRODUCTION
Mixed actinide dioxides, such as (U,Pu)O2, are already used as
fuels in pressurized water reactors (PWRs) (including Gen III,
EPR) and stand as potential fuels for several Gen IV concepts,
such as sodium-cooled fast reactor (SFR) or gas-cooled fast
reactor (GFR).1,2 Moreover, as the reprocessing of actinides
coming from nuclear spent fuel into new fuel elements has
emerged to economize uranium resources, to reduce the long-
term nuclear waste radiotoxicity, and to increase the resistance to
plutonium proliferation, the recycling of minor actinides, includ-
ing neptunium, americium, and curium, in such mixed-oxide fuels
or in UO2-based blankets surrounding the core is now often
considered.3
In this context, it appeared important to evaluate the con-
sequences of the modifications linked to the initial precipitation
of oxalate precursors on several properties of interest and,
particularly, on dissolution kinetics, which stands as a key
parameter for the reprocessing operations. Consequently, this
general study was undertaken to highlight the links between the
structure and the microstructure of Ce1ꢀxNdxO2ꢀx/2 model
dioxide materials and their ability to dissolve. Indeed, CeO2 is
frequently used as a surrogate for plutonium dioxide12 since both
oxides crystallize in the same fluorite-type structure (face-cen-
tered cubic; space group, Fm3m; JCPDS cards 01-081-0792 for
Mixed actinide dioxide fuels are currently prepared through
dry chemistry routes based on powder mixtures, which could
result in some heterogeneity in the cationic distribution in the
material.4 More recently, various studies were dedicated to the
preparation of homogeneous solids through wet chemistry
methods based on the precipitation of crystallized precursors.5
Among them, carbonates,6 nitrates,7 or hydroxides8 were re-
ported in the literature. However, oxalate compounds remain
probably the most frequently cited due to their quantitative
precipitation and to the good properties they confer to the final
solids obtained after heating (i.e., cationic distribution in the
structure, sintering capability, or chemical durability).9,10 Indeed,
CeO2 and 00-051-0798 for PuO2) with close unit cell parameters
VIII
(Table 1) due to similar ionic radii (
r
= 0.97 Å and VIIIrPu4+
=
Ce4+
0.96 Å13). Similarly, neodymium (VIIIrNd3+ = 1.109 Å) can be used
to mimic the behavior of trivalent minor actinides, such as
VIII
Am3+
r
americium (
= 1.09 Å) and curium.14ꢀ17
Some studies already showed that several parameters, such as
density, porosity, grain size, or density of grain boundaries, could
have a significant impact on the normalized dissolution rates.9,23
Received: June 13, 2011
Published: August 02, 2011
r
2011 American Chemical Society
9059
dx.doi.org/10.1021/ic201269c Inorg. Chem. 2011, 50, 9059–9072
|

Inorganic Chemistry
ARTICLE
Table 1. Unit Cell Parameter (a) of Dioxides Crystallized in
the Fluorite-Type Structure
thermodynamically favorable. The calculation of the free
enthalpy variation of the dissolution reaction without any
change in the oxidation state (eq 1) led to a positive variation
of the Gibbs energy:31
CeO2
VIIIr(cat) (Å)13 0.97
ThO2
UO2
NpO2
PuO2
AmO2
1.05
5.4104(2) 5.597(2) 5.469(1) 5.433(1) 5.395(1) 5.3736(5)
18 19 20 21 21 22
1.00
0.98
0.96
0.95
CeO2 þ 4Hþ T Ce4þ þ 2H2O
a (Å)
reference
ΔRG° ¼ 40:1 kJ molꢀ1
ð1Þ
3
After the preparation of a panel of powdered CeO2 and
Ce1ꢀxNdxO2ꢀx/2 samples, this work was thus mainly dedicated
to the influence of their crystallization state on their normalized
dissolution rate in acid solutions. Its relative promoting role was
compared to that of temperature or acid concentration.
2. EXPERIMENTAL SECTION
2.1. Synthesis of Cerium(IV)ꢀNeodymium(III) Oxide Solid
Solutions. Chemical reagents were supplied by VWR, and Aldrich-
Fluka, whereas cerium(III) or neodymium(III) chloride hexahydrates
were issued from Strem Chemicals (g99.9% purity). These latter were
dissolved in 0.5 M HNO3 to obtain solutions with a final concentration
of 0.1 M in cations. The synthesis of the initial oxalate precursors was
performed through a mixture of Ce (III) and Nd (III) solutions in the
desired stoichiometry (mole ratios of x = 0, 0.1, and 0.2) with a large
excess of oxalic acid at room temperature, under stirring. The quanti-
tative precipitation of the cations occurred immediately according to the
following chemical reaction:
ð2 ꢀ 2xÞCe3þ þ 2xNd3þ þ 3H2C2O4
þ yH2O f Ce2ꢀ2xNd2xðC2O4Þ3 3 yH2O þ 6Hþ
ð2Þ
Mixed oxalate precipitates were then washed several times with
deionized water, filtered, and dried overnight in an oven (90 °C).
Whatever the chemical composition considered, XRD patterns revealed
that the samples prepared were single phase and fitted well with all the
XRD lines associated with the lanthanide oxalate decahydrate Ln2-
(C2O4)3 10 H2O (Ln = LaꢀEr; monoclinic structure; space group,
3
P21/c).32,33
The transformation of (Ce,Nd)-oxalate into a mixed dioxide was then
followed through thermogravimetric analysis (TGA) and differential thermal
analysis (DTA) (Figure 1), performed on a Setsys Evolution apparatus
provided by Setaram. Heating treatments were performed in air, with a
constant rate of 5 °C minꢀ1 up to 300ꢀ1200 °C, identical to that
Figure 1. Results of TGA (a) and DTA (b) obtained when heating
Ce2ꢀ2xNd2x(C2O4)3 10H2O (x = 0, 0.09, and 0.16) in air with a heating
3
3
applied for the preparation of final oxides.
rate of 5 °C minꢀ1
.
2.2. Characterization. To evaluate the influence of the crystal-
lization state on several physicochemical properties of powdered
Ce1ꢀxNdxO2ꢀx/2 samples, oxalate precursors were fired for 2 h at
different temperatures ranging from 300 to 1200 °C, and the resulting
samples were extensively characterized.
However, they were poorly examined compared to the well-
known effects of temperature or pH in the literature. Indeed,
many studies were dedicated to the optimization of the operating
conditions for the dissolution of PuO224ꢀ26 even if the conclu-
sions on the mechanisms involved were far less numerous.
Nevertheless, they all demonstrated the refractory nature of
plutonium dioxide in both oxidative and reducing condi-
tions.27,28 In comparison, the dissolution of CeO2 was rela-
tively poorly examined. Cerium dioxide was generally re-
ported to be strongly resistant to corrosion in nitric acid
media. However, the formation of nitrate complexes promot-
ing the ceria dissolution was already mentioned. Indeed, few
studies reported the experimental conditions leading to the
quantitative dissolution of cerium dioxide and established the
associated kinetics.29,30 Moreover, the reaction of the dis-
solution of CeO2 in noncomplexing acid solutions is not
X-ray Powder Diffraction. XRD patterns were collected between 10
and 80° (θꢀ2θ mode) using a Bruker D8 Advance X-ray diffractometer
(Cu KR1,2 radiation, λ = 1.5418 Å) equipped with a linear Lynx-eye
detector. For all the powdered samples, a step of 0.01° and a counting
time of 1.5 s stepꢀ1 were considered. The unit cell parameters for
3
fluorite-type Ce1ꢀxNdxO2ꢀx/2 oxides were refined by the Rietveld
method using the ThomsonꢀCoxꢀHastings Pseudo Voigt convoluted
with an axial divergence asymmetry function34 available in the Full-
prof_Suite program.35
ESEM. In situ high-temperature environmental scanning electron
microscopy (HT-ESEM) experiments were performed using a field
emission gun environmental scanning electron microscope (model FEI
QUANTA 200 ESEM FEG) equipped with a 1500 °C hot stage. The
sample was directly placed in a 5 mm diameter MgO crucible covered
9060
dx.doi.org/10.1021/ic201269c |Inorg. Chem. 2011, 50, 9059–9072

Inorganic Chemistry
ARTICLE
Table 2. Expected and Experimental Compositions of
Fluorine-Type Ce1ꢀxNdxO2ꢀx/2 and Associated Refined Unit
Cell Parameters
released in the leachate were determined by inductively-coupled plasma
atomic emission spectroscopy (ICP-AES, Spectro Arcos) after dilution
of the samples with diluted nitric acid in order to reach a total volume of
analyzed solution of 6 mL. The intensity of the peaks was recorded at
λ = 448.691 and 413.765 nm for cerium and at λ = 406.109 and 430.358 nm
for neodymium to avoid any interference between the elements.
expected chemical formula
x (exptl)
unit cell
(Ce1ꢀxNdxO2ꢀx/2
)
(from EDS)
parameter (Å)
The normalized weight losses (NL, also called normalized leachings)
were calculated from the elementary concentrations after normalization
by the reactive surface S (m2) of the sample in contact with the solution
and by the weight stoichiometric ratio of the element considered in the
solid (fi, expressed as the ratio between the mass of the considered
element and the overall mass of the sample),38 following the equation
x = 0
0
a = 5.4107(2)
a = 5.4305(4)
a = 5.4489(1)
x = 0.1
x = 0.2
0.09
0.16
with platinum paint. The sample was heated directly in the ESEM
chamber with a rate of 20 °C minꢀ1 and considering several holding
mi
3
NLðiÞ ¼
ð3Þ
temperatures between room temperature and 1000 °C. The furnace
temperature was controlled by a thermocouple placed near the MgO
crucible, and the sample temperature accuracy was checked during each
experiment. Micrographs were recorded each 100 °C all during the heat
treatment of the sample on a selected part of the ceramic with different
magnifications. The observations were performed in air atmosphere at
an operating pressure of 120 Pa, and a specific detector was used for
in situ gaseous secondary electron imaging at high temperature.
The precise chemical composition of the samples was checked by the
means of energy dispersive spectrometry (EDS) coupled with the ESEM
device. For this purpose, each compound was previously shaped through
uniaxial pressure, sintered for 10 h at 1400 °C in air, then coated with an
epoxy resin, polished, and finally metalized by carbon deposition. EDS
analyses were then performed considering 12 data acquisitions. At least
40 000 counts spectra were collected by the EDS detector using an
optimal working distance of 11.4 mm and an acceleration voltage of
15 kV. In these conditions, the average xNd value was assumed with an
absolute accuracy of (0.01 since all the data acquisitions only led to a
weak dispersion.
fi ꢁ S
where mi (expressed in g) corresponds to the total amount of the
element i measured in the solution.
Using the approach described by Lasaga39 and derivating eq 3 as a
function of leaching time, the normalized dissolution rate of the element
i was determined from38
dNLðiÞ
dt
1
dmi
dt
RLðiÞ ¼
¼
ꢁ
ð4Þ
fi ꢁ S
In such conditions, the normalized dissolution rates correspond to
the slopes obtained when plotting the evolution of the normalized
weight losses. The normalized leachings NL (expressed in g mꢀ2) and
3
the associated normalized dissolution rates RL (expressed in g mꢀ2
d
3
3
ꢀ1) were determined considering the following equation
ꢀ
ꢁ
dNLðiÞ
dt
d
Ci ꢁ V ꢁ Mi
fi ꢁ S
V ꢁ Mi dCi
RLðiÞ ¼
¼
≈
ꢁ
ð5Þ
dt
fi ꢁ S
dt
where Ci is the concentration of the lanthanide in the leachate
(mol Lꢀ1), V the volume of leachate (L), fi the weight stoichiometric
The results of characterization of Ce1ꢀxNdxO2ꢀx/2 samples by
X-EDS are gathered in Table 2. From these results, all the samples were
found to be homogeneous and single phase. Only slight deviations in the
chemical compositions of Ce1ꢀxNdxO2ꢀx/2 were noted compared to
the calculated values.
Specific Surface Area and Porosity Distribution. The specific surface
area measurements were performed with a Micromeritics ASAP 2020
apparatus using nitrogen adsorption (Brunauer, Emmet, and Teller,
BET method) at 77 K. The pore size distribution of the oxide samples
were estimated from the desorption branches of the nitrogen sorption
isotherm according to the BJH (Barret, Joyner, and Halenda) method,
based on the Kelvin equation, which correlates the pore size with critical
condensation pressure, and by assuming a straight cylindrical pore
model (that is not always the case for all the studied materials).36 Prior
to the measurements, the samples were dried at 300 °C for 3 h to ensure
their complete outgassing.
2.3. Leaching Experiments. Leaching tests were performed using
batch experiments in polytetrafluoroethylene containers at room tem-
perature. For each dissolution experiment, about 125 mg of powder was
put in contact with 25 mL of 0.1ꢀ6 M HNO3 for a few weeks to several
months. After settling, an aliquot of 700 μL of the leachate was taken off,
then renewed with fresh solution at regular intervals to determine the
evolution of the elementary releases. In such conditions, less than 3% of
the leachate was removed in order to avoid any perturbation in the
establishment of saturation processes at the solid/liquid interface.
Aliquots were then centrifuged at 12 000 rpm to avoid the presence of
colloids smaller than 11 nm.37
3
ratio of the lanthanide considered in the solid (dimensionless) and Mi its
molar mass (g molꢀ1), and S the reactive surface of the solid (m2).
3
In this expression, the stoichiometric ratio was assumed to remain
almost constant during the leaching tests, which could not be system-
atically the case for all the ceramic materials. However, this assumption
was supported by the strong refractory character of CeO2,29,30 and by
the associated very low normalized dissolution rates determined, as
reported in the following sections. On the basis of an average normalized
dissolution rate of 10ꢀ5 g mꢀ2
would be dissolved after 100 days of leaching.
d
ꢀ1, less than 1% of the initial material
3
3
The normalized mass loss would increase linearly with time during
the first days of leaching tests. However, the real behavior was often
found rather different since an initial pulse in the elementary releases was
observed. This phenomenon, which remained limited for these studied
materials, was usually assigned to the presence of crystal defects,
nonstoichiometric phases, minor phases, or small particles at the surface
of the unwashed samples.38,40,41 It usually produced large amounts of
colloids during the first days of leaching tests. To avoid this problem, an
initial washing of the samples was performed for 3 days in 2 M HNO3 at
room temperature, before the beginning of the dissolution experiments.
Moreover, for longer leaching times (i.e., near the saturation conditions
in the leachate), the precipitation of neoformed phases usually induced
the formation of a protective layer acting as a diffusion barrier for the
released elements, leading to the significant decrease of the associated
normalized dissolution rates.
Thereafter, the reaction of dissolution is called stoichiometric if all the
elements i are released in the leachate with the same normalized
dissolution rate. Conversely, if one element is released more rapidly
from the solid (e.g., like fluorine in the britholite structure40), the
dissolution is called selective regarding to the considered element.
Moreover, the dissolution is qualified as congruent when all the
Furthermore, the behavior of Ce1ꢀxNdxO2ꢀx/2 samples during
dissolution processes was also examined through leaching tests per-
formed with high leachate renewals (1 mL hꢀ1), called “dynamic”
3
experiments (2 M HNO3, T = 90 °C), in order to avoid any saturation
processes. Finally, the concentrations of cerium and neodymium
9061
dx.doi.org/10.1021/ic201269c |Inorg. Chem. 2011, 50, 9059–9072

Inorganic Chemistry
ARTICLE
Table 3. Partial Order Related to the Proton Activity and
Activation Energy Obtained during the Dissolution of Several
Ceramics, Minerals, and Materials
(2) Reaction of adsorbed species among themselves or with atoms of
the surface.
(3) Desorption of the product species formed at the surface.
The last reaction is usually slower and, therefore, controls the overall
rate of dissolution of the solid.
partial
EA
mineral/material
order n
reference (kJ molꢀ1) reference
3
albite (NaAlSi3O8)
0.49
50
53
48
47
48
57
48
47
59
27
71
56
38
3. RESULTS AND DISCUSSION
β-TPD (Th4(PO4)4P2O7) 0.31ꢀ0.35
ThO2
42 ( 3
3.1. Preparation of Ce1ꢀxNdxO2ꢀx/2 Solid Solutions. The
final oxides were obtained after heating the lanthanide oxalate
0.26 ( 0.05
0.41 ( 0.05
0.91 ( 0.09
0.53
20 ( 3
35 ( 3
47
58
47
27
decahydrates Ln2(C2O4)3 10H2O in air in alumina boats,
UO2
3
according to the following chemical reaction:
Th0.63U0.37O2
0.30 ( 0.01
0.55 ( 0.10
0.50 ( 0.06
1.7
Δ
CeI2IIꢀ 2xNdI2IxIðC2O4Þ3 yH2O f 2Ce1IVꢀ xNdxIIIO2ꢀx=2
33 ( 4
NDa
3
þ ð4 ꢀ xÞCO v þ ð2 þ xÞCO2 v þ yH2Ov
ð8Þ
Th0.87Pu0.13O2
U0.75Pu0.25O2
a ND: not determined.
10ꢀ32
This chemical transformation was characterized by a strong
change in the powder color from white to yellow, due to the
oxidation of cerium(III) into cerium(IV). The chemical trans-
formation of Ce2ꢀ2xNd2x(C2O4)3 10H2O into Ce1ꢀxNdx-
O2ꢀx/2 was then followed through TGA/DTA experiments
(Figure 1).
normalized dissolution rates are identical (i.e., when all the elements are
released with the same ratios than the stoichiometry of the initial
material). Conversely, it is called incongruent if at least one element is
precipitated as neoformed phases in the back end of the initial reaction of
dissolution. More generally, the dissolution is described as congruent for
1/3 < Ri/Rj < 3.41,42
Influence of Acidity. Many authors already demonstrated that the
normalized dissolution rates of most of the minerals increased with the
proton concentration for pH < 7.41ꢀ48 Indeed, in acid solutions, the RL
values were found to be proportional to a fractional power of the proton
activity as follows49ꢀ53
3
Whatever the chemical composition examined, the variation of
the weight loss versus heating temperature exhibited two steps
corresponding to a total loss of 53%, close to that calculated
(52.5%). The first one (weight loss of 25%) occurred between 20
and 200 °C and was correlated to the loss of 10 water molecules,
leading to the formation of anhydrous Ce2ꢀ2xNd2x(C2O4)3. The
decomposition of oxalates occurred between 280 and 360 °C
with the additional loss of CO and CO2 molecules (experimental
mass loss of 26%; calcd: 27%). This result was in good agreement
with the observations made by several authors on the decom-
position of cerium oxalate.60,61 Conversely to what was observed
during the thermal decomposition of mixed Ce and Gd
oxalates,60,61 the formation of carbonated or oxo-carbonated
intermediates was not clearly evidenced in this study. Indeed, the
data reported in Figure 1B indicate that the Ce2ꢀ2xNd2x-
n
n
RL ¼ k0T ꢁ ðH3OþÞ ¼ k0T ꢁ ðγ
ꢁ ½H3OþꢂÞ
ð6Þ
H3Oþ
where RL refers to the proton-promoted normalized dissolution rate and
k0T (expressed in g mꢀ2 dayꢀ1) represents the apparent normalized
3
3
dissolution rate constant at ꢀlog([H3O+]) = 0. k0T is independent of the
acidity of the leachate, but temperature-dependent. The n parameter
corresponds to the partial order related to the proton activity, (H3O+) to
the proton activity, and γH3O+ corresponds to the proton activity
coefficient. In this study, γH3O+ values were calculated according to the
Pitzer and SIT models.54 The values of partial orders related to protons,
n (and to hydroxide ions in basic media), were generally found between
0 and 1 for several materials and minerals (Table 3), suggesting that the
dissolution was mainly driven by surface reactions occurring at the
solidꢀliquid interface.
(C2O4)3 10H2O precipitates were fully converted to Ce1ꢀx
-
3
NdxO2ꢀx/2 solid solutions when heating above 400 °C. More-
over, no endothermic peak associated with the formation of
carbonate or oxocarbonate was observed prior to the formation
of oxide. This result was also checked by in situ μ-Raman
spectroscopy experiments on several lanthanide oxalates.
Moreover, the exothermic peak observed in the DTA curves
and associated with the decomposition of oxalate entities (thus,
to the release of CO and CO2) was measured at higher
Influence of Temperature. Temperature is also an important para-
meter affecting the normalized dissolution rate according to the
Arrhenius law, that is
temperature for Ce2ꢀ2xNd2x(C2O4)3 10H2O samples than for
3
n
Ce2(C2O4)3 10H2O. This result was consistent with the data
RL ¼ k ꢁ eꢀðE =RTÞ ¼ k00 ꢁ ðH3OþÞ ꢁ eꢀðE =RTÞ
ð7Þ
A
A
3
reported all along the Ce2ꢀ2xNd2x(C2O4)3 10H2O series,62
3
where the temperature slightly increased when increasing the
neodymium content (T = 310, 370, and 410 °C for x = 0, 0.285,
and 0.59, respectively). It also agreed well with the data reported
by Higashi et al.61 and Sharov et al.,63 who showed an increase of
the decomposition temperature with the ionic radius of the rare-
earth element during the decomposition of Ce2ꢀ2xLn2x-
where k is the normalized dissolution rate constant independent of
temperature (expressed in g mꢀ2 dayꢀ1) and EA is the apparent
3
3
activation energy of the dissolution of the mineral (expressed in kJ
molꢀ1). Plotting the variation of ln(RL) versus the reciprocal tempera-
ture allows determining the activation energy and associated k value.
Because of the formation of adsorbed species onto the leached surface,
such EA values were usually found to be significantly lower than the
energy of formation of chemical bonds (Table 3).
Most of the available theories indicate that the reaction of dissolution
is controlled by the decomposition of an activated complex55 according
to a three-step process:
(C2O4)3 10H2O (Ln = Yb, Gd, Sm, Nd, La). The small
3
difference observed between both solid solutions (ΔT =
15 °C) could result from the uncertainties associated with the
temperature determination but could also arise from the small
differences observed in the initial powder reactivity (SBET = 2 and
(1) Adsorption of aqueous species onto the surface.
5 m2 gꢀ1 for x = 0.09 and x = 0.16, respectively).
3
9062
dx.doi.org/10.1021/ic201269c |Inorg. Chem. 2011, 50, 9059–9072

Inorganic Chemistry
ARTICLE
Figure 2. XRD patterns of CeO2 recorded after heating the samples at
various temperatures (air, 2 h). All the XRD lines of the sample holder
are pointed out by stars (/).
Figure 3. Variation of the average crystallite size versus the heating
temperature obtained for CeO2 (red circle), Ce0.91Nd0.09O1.955 (black
square), and Ce0.84Nd0.16O1.92 (blue triangle) samples.
3.2. Structural and Microstructural Characterization. To
evaluate the influence of the crystallization state on the behavior
of Ce1ꢀxNdxO2ꢀx/2 samples during leaching tests, the structural
(and microstructural) characterization of the powdered samples
versus heating temperature was undertaken in terms of crystal-
lization state (XRD), morphology (SEM), and specific surface
area (N2 adsorption-BET method).
corundum reference. As expected from the peak shapes, the
FWHM value strongly decreased from 300 to 600 °C, then
progressively stabilized for T > 800 °C. The crystallization
state of CeO2 was considered to be optimized above this
temperature.
In a second step, the application of the Rietveld method
allowed determining the size of coherent domains (i.e., the
average crystallite size) corresponding to the various crystal-
lization states assumed from the FWHM values (Figure 3).
Between 300 and 700 °C, they were found to vary from 5 to
40 nm (CeO2) and from 5 to 15 nm (Ce1ꢀxNdxO2ꢀx/2),
respectively. For higher temperatures, the crystallite size strongly
XRD characterization. The Ce2ꢀ2xLn2x(C2O4)3 10H2O pre-
3
cursors were thus heated each 100 °C from 200 to 1200 °C, then
characterized by XRD (Figure 2). At 200 °C, the XRD patterns
did not evidence any XRD lines, showing that the samples were
fully amorphous. This observation was correlated to the full
dehydration of the initial hydrated oxalates, as shown from DTA
and TGA experiments. It appeared in good agreement with the
results of Balboul et al. on the amorphous nature of Nd2-
(C2O4)3.64 All the XRD lines characteristic of the fluorite-type
structure adopted by Ce1ꢀxNdxO2ꢀx/2 mixed dioxides for x <
0.462,65 were then observed above 300 °C.66 Moreover, no
additional phase was detected, showing that trivalent neody-
mium was inserted in the structure in the cerium(IV) site, leading
to the formation of Ce1ꢀxNdxO2ꢀx/2 solid solutions.67 As
already described in the literature, this substitution was asso-
ciated with the formation of oxygen vacancies to ensure the
charge compensation.62,68,69 From 300 to 600 °C, the XRD lines
remained broad and indicated that the crystallization state was
not optimal. On the contrary, it was clearly improved between
700 and 1200 °C (where the XRD lines were found to be strongly
narrowed). The associated refined unit cell parameters of
CeO2, Ce0.91Nd0.09O1.955, and Ce0.84Nd0.16O1.92 reached a =
5.4107(2), 5.4305(4), and 5.4489(1) Å, respectively (Table 2).
All these values were in good agreement with the literature (a =
increased: it reached 260, 140, and 80 nm for CeO2, Ce0.91
-
Nd0.09O1.955, and Ce0.84Nd0.16O1.92, respectively, when heating
the samples at 1200 °C. This observation suggested a two-step
process in the growth of crystallite size. It was first explained by
the elimination of crystal defects and amorphous domains (T e
700 °C), leading to the improvement of the crystallization state
of the solid associated with a limited growth of the coherent
domains. The second step, observed for temperatures higher
than 800 °C, was mainly associated with the growth of crystallites
through diffusion phenomena close to that responsible for
sintering processes at the grain scale for higher temperatures. It
suggested an Ostwald ripening process where small crystals were
progressively integrated to form bigger ones that are stable at
high temperatures.70,71
The important difference in the crystallite size observed
between the three samples after heating at 1100 °C directly
resulted from the influence of the neodymium incorporation in
the structure. Such an effect was correlated to that observed
during the DTA/TGA experiments from which a shift to the
higher temperatures of all the chemical transformations (e.g., for
the oxalate to oxide transformation) occurred when increasing
the neodymium content in the structure. From the literature, this
effect was mainly assigned to an heterogeneous distribution of
neodymium in the sample. In this sense, Belliꢀere et al. evidenced
an enrichment in trivalent lanthanides within the grain bound-
aries and at the surface due to the space charge effect provoked by
the lower charge of Ln3+ compared to that of Ce4+.72 They
argued that the excess of trivalent cations located within the
crystallite boundaries generated a gradient in concentration
5.41134(8) Å for CeO2, a = 5.43022(8) Å for Ce0.9Nd0.1O1.95
,
and a = 5.45084(11) Å for Ce0.8Nd0.2O1.918,62). This increase
with the trivalent neodymium mole loading was consistent with
VIII
the incorporation of the bigger Nd(III) (
r = 1.109 Å)
Nd3+
VIII
compared to Ce(IV) (
r
= 0.97 Å).
Ce4+
To follow more accurately the variation of the crystallization
state of Ce1ꢀxNdxO2ꢀx/2 solid solutions versus the heating
temperature, the average full width at half-maximum (FWHM)
was determined from the seven more intense XRD lines for all
the prepared samples and were compared to that of the
9063
dx.doi.org/10.1021/ic201269c |Inorg. Chem. 2011, 50, 9059–9072

Inorganic Chemistry
ARTICLE
crystallite size, and EA,CG is the activation energy for crystallite
growth. The plot of log(D) versus the reciprocal temperature led
to an activation energy of 21 ( 2 kJ molꢀ1 for the three
3
compounds and for 700 °C e T e 1200 °C. Such a value was
found to be independent of the chemical composition and was in
good agreement with the experimental results obtained for
nanopowdered samples of CeO2 (EA,CG = 17 kJ molꢀ1).74
3
Morphological Characterization. The specific surface area of
the prepared samples was followed versus the heating tempera-
ture (Figure 4A) and coupled to ESEM observations (Figure 5).
Whatever the sample studied, the specific surface area reached
a maximum value higher than 80 m2 gꢀ1 when heating the
3
samples between 300 and 400 °C. Such an increase was already
reported by several authors for the decomposition of Th1ꢀxUx-
(C2O4)2 2H2O10 or Ce2(C2O4)3 10H2O75 oxalates. This in-
3
3
crease was usually correlated to the dehydration of the samples,
then to the decomposition of the oxalate entities into CO and
CO2, which induced the formation of defects (cracks, pores...)
through the establishment of strong constraints associated with
the release of large amounts of gaseous molecules. This range of
temperatures was consistent with that expected from TGA and
DTA experiments. Indeed, the oxalate-to-oxide transformation
started at 300 and 350 °C for CeO2 and Ce1ꢀxNdxO2ꢀx/2
,
respectively. Moreover, the observation of such a phase trans-
formation and the subsequent formation of defects leading to the
increase of the specific surface area was also correlated to the
XRD data that revealed poorly crystallized samples in this range
of temperatures.
Above this temperature, the specific surface area progressively
decreased as a consequence of the improvement of the crystal-
lization states of the samples and of crystallite growth processes.
The decrease of the specific surface area was faster for CeO2 than
for Ce1ꢀxNdxO2ꢀx/2 (14 and 37 m2 gꢀ1, respectively, for a
3
heating temperature of 600 °C). This result was directly con-
nected to the variation of the crystallinity of the resulting
samples. At this temperature, the FWHM was twice larger for
Ce1ꢀxNdxO2ꢀx/2 compared with CeO2. This difference was kept
up to 1000 °C, where the specific surface area decreased to 7 and
2 m2 gꢀ1 for Ce0.91Nd0.09O1.955 and CeO2, respectively. Simul-
3
taneously, the average crystallite size increased up to 40 nm
(Ce0.91Nd0.09O1.955) and 170 nm (CeO2). Because the morphol-
ogy of the powder was mainly conserved, such variations of the
specific surface area were attributed to porosity modifications,
consequently to the release of gaseous molecules during the
oxalate conversion. Desorption isotherm analysis of Ce0.91
-
Nd0.09O1.955 and Ce0.91Nd0.09O1.92 with the BJH method was thus
performed in order to give additional information on the pore
size distribution (Figure 4B,C). As expected from BET experi-
ments, the modifications in the pore distribution were strongly
affected by the heating temperature in the same way for both
solid solutions. Indeed, the samples prepared at 400 °C exhibited
nanometric porosity (with a pore size ranging from 3.5 to
4.5 nm) associated with ≈80% of the developed specific surface
area. Because such nanometric porosity was not observed for the
Ce0.91Nd0.09O1.955 samples prepared at 100 and 200 °C, it was
mainly associated with the decomposition of oxalate into oxide
(and thus to the release of large amounts of gaseous molecules).
For Ce0.91Nd0.09O1.955, this pore size population remained
observable up to 800 °C even if the associated surface area
developed decreased significantly from 400 to 800 °C. Moreover,
its relative contribution to the global surface remained predomi-
nant and almost constant (79ꢀ91% of SBET) between both
Figure 4. Variation of the specific surface area versus the heating
temperature (t = 2 h) observed for CeO2 (red circle), Ce0.91Nd0.09O1.955
(black square), and Ce0.84Nd0.16O2 (blue triangle) samples (A) and
variation of the incremental pore area versus the average diameter of
pores for Ce0.91Nd0.09O1.955 heated at 100, 200, 300, 400, 500, 600, 800,
and 1000 °C (B) and for Ce0.84Nd0.16O1.92 heated at 300, 400, and
500 °C (C).
between the bulk and the surface, curbing the crystallite bound-
ary migration and thus delaying the crystallites' growth.
The activation energy for the crystallite growth of CeO2 and
Ce1ꢀxNdxO2ꢀx/2, was estimated by eq 973
ꢀ
ꢁ
EA;CG
D ¼ k ꢁ exp ꢀ
ð9Þ
R ꢁ T
where k is a constant value, R the perfect gas constant, D the
9064
dx.doi.org/10.1021/ic201269c |Inorg. Chem. 2011, 50, 9059–9072

Inorganic Chemistry
ARTICLE
Figure 5. In situ HT-ESEM micrographs of the Ce2(C2O4)3 10H2O starting precursor (A) and of resulting CeO2 powders obtained after heating at
3
300 (B), 500 (C), 700 (D), 800 (E), and 1000 °C (F).
temperatures. After heating the Ce0.91Nd0.09O1.955 samples at
1000 °C, this nanometric porosity was significantly reduced
while the pore size increased up to 4.5ꢀ5.5 nm. This increase
was mainly explained by the coalescence of nanometric pores
simultaneous to the grain growth.
between 90 and 500 °C. This decrease in size of about 25% was
mainly associated with the release of gaseous H2O, CO, and CO2
during the oxalate-to-oxide transformation. The grain sizes
remained almost constant between 500 and 800 °C.
However, these micrographs did not explain the strong
variations of surface area between 200 and 800 °C because the
pores were too small (around 4 nm) to be highlighted unam-
biguously. At 1000 °C, the size of the aggregate was 5.9 μm in
length and was 1.6 μm in width; this further decrease was
explained by the beginning of the sintering of the powder.
3.3. Dissolution of CeO2 and Ce1ꢀxNdxO2ꢀx/2 Samples. As
already mentioned, the chemical durability of Ce1ꢀxNdxO2ꢀx/2
The decomposition of Ce2(C2O4)3 10H2O was also followed
3
in situ by HT-ESEM (Figure 5). ESEM micrographs showed that
the samples were constituted by nanometric grains associated in
lamellar-shaped aggregates of 2ꢀ10 μm in length. This mor-
phology was kept all during the oxalate-to-oxide conversion
(Figure 5AꢀC). However, the oxalate decomposition was
characterized by an isotropic contraction of the aggregate size
9065
dx.doi.org/10.1021/ic201269c |Inorg. Chem. 2011, 50, 9059–9072

Inorganic Chemistry
ARTICLE
Figure 6. Evolution of the normalized leaching (A) and of the cumulative normalized leaching (B) during leaching tests of Ce0.91Nd0.09O1.955 (fired at
1000 °C) in dynamic conditions (2 M HNO3, T = 90 °C): NL(Ce) (black square) and NL(Nd) (red circle).
samples prepared at 1000 °C was examined through leaching
tests performed in 2 M HNO3 at 90 °C with a leachate renewal of
1 mL hꢀ1 (called “dynamic” experiments) to avoid the presence
3
of saturation processes (Figure 6A,B). This kind of experiment
also allowed evaluating the initial normalized dissolution rates
(see section 2.3).41,47,48,76 The evolution of the instantaneous
normalized leachings NL(Nd) and NL(Ce) revealed a slight
preferential release of neodymium during the first days of
leaching that could reflect some differences in the energy of
cohesion of both trivalent neodymium and tetravalent cerium
within the samples. However, this phenomenon was mainly
observed during the first days of leaching when using dynamic
conditions. It was directly connected to the enrichment in
trivalent lanthanide within the grain boundaries and at the
surface of the samples already reported by Belliꢀere et al.72 Indeed,
on the basis of this surface enrichment in neodymium compared
to the bulk composition, the associated normalized dissolution
rates were expected to be higher and thus would be directly
correlated to this small difference between both elements during
the first days of leaching time. On the basis of the direct
comparison of both lanthanides, this initial pulse was always
lower than 5% and 1% of initial neodymium and cerium contents,
respectively. Consequently, the stoichiometric ratio fi was af-
fected by less than 3.3%, supporting the assumption made on its
constancy (eqs 4 and 5).
Figure 7. Evolution of the normalized leaching during leaching tests of
Ce0.91Nd0.09O1.955 (fired at 1000 °C) in static conditions (C) (2 M
HNO3, T = 60 °C): NL(Ce) (black square) showing the two trends (far
and near saturation conditions) usually observed during the dissolution
of ceramic materials.41,42,47,48
.
dissolution rates determined led to higher uncertainties asso-
ciated with the neodymium concentration compared to that of
cerium.
To evidence the effect of several parameters, such as nitric acid
concentration, leaching temperature, and crystallization state, the
dissolution of samples was also studied with low leachate renew-
als (called “static” experiments). This kind of experiment allowed
determining the initial normalized dissolution rates first, then
studying the establishment of saturation processes (see section
2.3).42 The variation of the normalized weight loss, NL(Ce),
determined when leaching Ce0.91Nd0.09O1.955 (initially fired at
1000 °C) in 2 M HNO3 at 60 °C is presented, as an example, in
Figure 7. The general evolution of NL(Ce) versus the leaching
time clearly exhibited two tendencies. The first one, which is
kinetically controlled, was observed before 10 days of leaching
time. This linear variation allowed determining the initial nor-
malized dissolution rate (RL,0(Ce) = (1.1 ( 0.2) ꢁ 10ꢀ4
As clearly shown in Figure 6A, the difference between both
elements (through the determination of the NL(Nd)/NL(Ce)
ratio) decreased when increasing the leaching time. Conse-
quently, the final corresponding cumulative normalized leachings
NL(Nd) and NL(Ce) varied linearly with rather similar slopes
(Figure 6B). The initial RL,0(Nd)/RL,0(Ce) ratio was initially
equal to 2.7 during the first day of leaching, then decreased down
to 1.3 after 6 days, where the associated normalized dissolution
rates RL,0(Ce) and RL,0(Nd) reached (6.4 ( 0.2) ꢁ 10ꢀ4 and
(7.8 ( 0.2) ꢁ 10ꢀ4 g mꢀ2
d
ꢀ1, respectively. Despite this initial
3
3
release, the dissolution was rapidly found to be congruent. This
point was confirmed for all the other leaching conditions
examined. Even if the release of both elements was followed,
the general trends of the normalized dissolution rate variation
were privileged on cerium since (1) it was the major component
of the leached samples and (2) the very slow normalized
g mꢀ2 ꢀ1). It was followed by a significant slope drop that
d
3
3
was usually assumed to be partly affected by the approach of
saturation conditions, allowing the determination of the
9066
dx.doi.org/10.1021/ic201269c |Inorg. Chem. 2011, 50, 9059–9072

Inorganic Chemistry
ARTICLE
Table 4. Initial Normalized Dissolution Rates RL,0(Ce) and RL,0(Nd) (g mꢀ2
Ce0.91Nd0.09O1.955 and CeO2 (Fired at 1000 °C) Performed at 60 °C in Various Nitric Acid Leaching Solutions
d
ꢀ1) Obtained during Leaching Tests of
3
3
compound
10ꢀ1 M HNO3
5 ꢁ 10ꢀ1 M HNO3
2 M HNO3
6 M HNO3
CeO2
Ce0.91Nd0.09O1.955
RL,0(Ce)
RL,0(Ce)
RL,0(Nd)
(3.2 ( 0.2) ꢁ 10ꢀ6
(4.1 ( 0.2) ꢁ 10ꢀ6
(1.0 ( 0.9) ꢁ 10ꢀ5
(8.6 ( 0.3) ꢁ 10ꢀ6
(3.9 ( 0.3) ꢁ 10ꢀ5
(2.0 ( 1.0) ꢁ 10ꢀ5
(2.9 ( 0.3) ꢁ 10ꢀ5
(1.8 ( 0.2) ꢁ 10ꢀ4
(3 ( 2) ꢁ 10ꢀ4
(3.3 ( 0.4) ꢁ 10ꢀ5
(3.1 ( 0.2) ꢁ 10ꢀ4
(8 ( 2) ꢁ 10ꢀ4
normalized dissolution rates near the establishment of thermo-
dynamic equilibria (RL,t(Ce) = (1.7 ( 0.2) ꢁ 10ꢀ5 g mꢀ2
d
ꢀ1).
3
3
Influence of the Nitric Acid Concentration. To underline the
influence of nitric acid concentration on the normalized dissolu-
tion rates, some leaching tests were performed in various media
(10ꢀ1, 5 ꢁ 10ꢀ1, 2, and 6 M HNO3). The initial normalized
dissolution rates RL,0(Ce) and RL,0(Nd) are gathered in Table 4,
while the variations of log RL(Ce) versus the leachate acidity are
plotted in Figure 8 for Ce0.91Nd0.09O1.955 and CeO2 (fired at
1000 °C). For all the media studied, the normalized dissolution
rates obtained when leaching CeO2 were very slow (from RL,0
-
(Ce) = (3.2 ( 0.2) ꢁ 10ꢀ6 g mꢀ2
d
ꢀ1 to RL,0(Ce) = (3.3 (
3
3
0.4) ꢁ 10ꢀ5 g mꢀ2
d
ꢀ1). Such values confirmed the high
3
3
chemical durability of this material during leaching or dissolution
experiments, as already described for powdered and sintered
ThO2 samples (RL,0(Th) = (1.7 ( 0.1) ꢁ 10ꢀ7 g mꢀ2 dꢀ1
3
3
when leaching the samples in 5 M HNO3 at T = 25 °C47). It also
suggested that no reduction of Ce(IV) into Ce(III) occurred
during the dissolution processes. Moreover, the behavior of the
samples was strongly affected by the nitric acid concentration, as
already described for ThO2 and Th1ꢀxUxO2 samples.47,48 In-
deed, an increase of 1 order of magnitude was obtained between
10ꢀ1 M HNO3 and 2 M HNO3 for CeO2 (RL,0(Ce) = (3.2 (
Figure 8. Variation of log(RL,0(Ce)) versus the opposite log(H3O+)
during leaching tests of Ce0.91Nd0.09O1.955 (black square) and CeO2
(red circle) in various HNO3 solutions (T = 60 °C).
dissolution of NaAlSi3O8 showed thatthen valuewas independent
of the central charge of the cation.52 The authors suggested that
this partial order could be directly connected to the number of
protons adsorbed onto the surface of the leached samples and thus
to the concentration of active sites at the solid/liquid interface.
From the cerium releases, the n values reached 0.63 ( 0.05 for
CeO2 and 1.1 ( 0.1 for Ce0.91Nd0.09O1.955, showing different
behaviors for the two solids during leaching tests. In the case of
Ce0.91Nd0.09O1.955, the n value was in good agreement with that
determined from the neodymium releases (1.1 ( 0.3) despite a
high associated uncertainty mainly due to the low concentrations
of neodymium measured in the leachate. The value obtained for
CeO2 was consistent with that reported for the other materials
and minerals, including the refractory ThO2 (Table 3). Because
n > 1 for Ce0.91Nd0.09O1.955, the dissolution of Ce1ꢀxNdxO2ꢀx/2
solid solutions may be mainly governed by the presence of
oxygen vacancies (and, consequently, strongly affected by the
decrease of the energy of cohesion), associated with the incor-
poration of a trivalent element in the structure instead of by the
adsorption of active species onto the surface of the sample.
Indeed, the rates associated with the elementary releases from
these solids would be significantly higher than that of adsorption
species at the solid/liquid interface. The same conclusions were
made for uranium-enriched Th1ꢀxUxO2 solid solutions when
comparing the uranium oxidation rate with that of detachment of
species from the solid by complexion.48 The chemical durability
of Ce1ꢀxNdxO2ꢀx/2 solid solutions would be strongly affected by
the M(IV)/M(III) cationic substitution when increasing the
neodymium content.62
0.2) ꢁ 10ꢀ6 and (2.9 ( 0.4) ꢁ 10ꢀ5 g mꢀ2
d
ꢀ1, respectively)
3
3
and almost 2 orders of magnitude for Ce0.91Nd0.09O1.955
(RL,0(Ce) = (4.1 ( 0.2) ꢁ 10ꢀ6 and (1.8 ( 0.2) ꢁ 10ꢀ4
g mꢀ2 ꢀ1, respectively). From these results, the behavior of
d
3
3
the Ce0.91Nd0.09O1.955 solid solution was thus more affected than
CeO2 by the increase of nitric acid concentration. This difference
surely resulted from the presence of oxygen vacancies in the
mixed ceriumꢀneodymium oxide that led to a weakening of the
crystal lattice and thus to a decrease of the chemical durability of
the leached samples.62
Moreover, as described by Lasaga,39 the normalized dissolu-
tion rate should increase linearly with the logarithm of the proton
activity at the solidꢀsolution interface. That was observed for the
major part of the nitric acid solutions considered. However, we
observed a decrease of the influence of nitric acid concentration
for 6 M HNO3 (Figure 8). In such conditions, increasing the
nitric acid concentration probably led to the progressive satura-
tion of active sites present at the surface of the samples that
reduced significantly the effect of protons in this range of
concentration.
As expected from eq 7, the linear variation of log(RL,0) versus
log(H3O+) allowed determining the partial order related to the
proton activity, n, for both CeO2 and Ce0.91Nd0.09O1.955 samples.
The free proton concentrations were determined according to
the literature.77,78 The fractional dependence with proton activity
was usually interpreted in several ways according to the authors.
The study of Al2O3 and BeO dissolutions already evidenced
that the partial order related to the protons activity was equal to
the charge of the central cation.51 However, the study of the
Influence of the Temperature. As already discussed, tempera-
ture is also a key parameter affecting the behavior of samples
9067
dx.doi.org/10.1021/ic201269c |Inorg. Chem. 2011, 50, 9059–9072

Inorganic Chemistry
ARTICLE
during dissolution. During leaching experiments of CeO2 in 2 M
HNO3, the normalized dissolution rate was increased by a factor
of 2 when increasing the leaching temperature from 60 °C
interface, as already described in the literature,38,39,47,52 and were
too high to consider that only diffusion phenomena controlled
the dissolution (since EA > 20 kJ molꢀ1).79
(RL,0(Ce) = (2.9 ( 0.3) ꢁ 10ꢀ5 g mꢀ2
d
ꢀ1) to 90 °C
Influence of the Crystallization State. Some rare and recent
publications dealt with the study of discrepancies observed in the
ThO2 solubility.80 They mainly underlined a lack of thorough
solid-state analysis and pointed out the role of specific site
exchange mechanisms at the solid/liquid interface through
experiments based on isotopic exchanges. However, the studies
devoted to the links observed between physicochemical proper-
ties of the samples and their behavior during leaching tests were
rarely considered,7 while they probably exhibited an important
contribution on the normalized dissolution rates. For this reason,
the effect of additional parameters on the normalized dissolution
rates, such as the crystallization state of the samples, was also
ꢀ3 2 ꢀ31
(RL,0(Ce) = (5.5 ( 0.4) ꢁ 10ꢀ5 g m
d ). For Ce0.91N-
3
3
d0.09O1.955, the increasing factor was equal to 2.5 (RL,0(Ce) =
(1.8 ( 0.2) ꢁ 10ꢀ4 g mꢀ2
d
ꢀ1 for T = 60 °C and (4.5 ( 0.2) ꢁ
3
3
10ꢀ4 g mꢀ2
d
ꢀ1 for T = 90 °C). According to eq 7, the slopes
3
3
obtained by linear regression when plotting log(RL,0) versus the
reciprocal temperature for both CeO2 and Ce0.91Nd0.09O1.955
samples led to an apparent activation energy between 36 and
37 kJ molꢀ1. Such values were consistent with the control of the
dissolution by surface reactions occurring at the solid/liquid
examined. The chemical durability of powdered CeO2, Ce0.91
-
Nd0.09O1.955, and Ce0.84Nd0.16O1.92 samples with various crystal-
lization states was studied in 2 M HNO3 at 60 °C. For instance,
the evolution of the normalized weight loss, NL(Ce), obtained
when leaching powdered CeO2 samples heated at various
temperatures between 300 and 1000 °C, is gathered in Figure 9.
For the three chemical compositions examined, the resulting
initial normalized dissolution rates RL,0(Ce) and RL,0(Nd) are
gathered in Table 5, while the variations of RL,0(Ce) versus the
temperature of heating treatment are reported, as an example, in
Figure 10. For all the heating temperatures considered, the more
important the neodymium content in the solid solution, the more
rapid the normalized dissolution rate, as the proof of the
important role of oxygen vacancies on the chemical durability
of the samples.
Despite the normalization of the elementary releases by the
mass ratios (fi) and by the reactive surface area (S) associated
with the solid/liquid interface (see eqs 3 and 4), the initial
normalized dissolution rates, RL,0(Ce) and RL,0(Nd), decreased
Figure 9. Evolution of the normalized leachings NL(Ce) obtained
during the dissolution of powdered CeO2 samples (2 M HNO3,
T = 60 °C) prepared for various temperatures of heating treatment.
Table 5. Initial Normalized Dissolution Rates RL,0(Ce) and RL,0(Nd) (Expressed in g mꢀ2 ꢀ1) Obtained during the Dissolution
d
3
3
of CeO2, Ce0.91Nd0.09O1.955, and Ce0.84Nd0.16O1.92 for Several Calcination Temperatures (2M HNO3, T = 60 °C)
heating temperature
300 °C
400 °C
500 °C
600 °C
CeO2
RL,0(Ce)
RL,0(Ce)
RL,0(Nd)
RL,0(Ce)
RL,0(Nd)
(1.1 ( 1) ꢁ 10ꢀ4
(5.1 ( 0.2) ꢁ 10ꢀ4
(1.2 ( 0.1) ꢁ 10ꢀ3
(1.2 ( 0.1) ꢁ 10ꢀ3
(2.5 ( 0.3) ꢁ 10ꢀ3
(6.7 ( 1.5) ꢁ 10ꢀ5
(4.1 ( 0.3) ꢁ 10ꢀ4
(9.3 ( 0.9) ꢁ 10ꢀ4
(9.3 ( 0.7) ꢁ 10ꢀ4
(2.4 ( 0.3) ꢁ 10ꢀ3
(5.1 ( 1.2) ꢁ 10ꢀ5
(2.3 ( 0.2) ꢁ 10ꢀ4
(6.9 ( 0.9) ꢁ 10ꢀ4
(6.0 ( 0.8) ꢁ 10ꢀ4
(1.3 ( 0.2) ꢁ 10ꢀ3
(5.0 ( 1.1) ꢁ 10ꢀ5
(1.6 ( 0.2) ꢁ 10ꢀ4
(4.1 ( 0.5) ꢁ 10ꢀ4
(7.0 ( 0.5) ꢁ 10ꢀ4
(6.2 ( 0.5) ꢁ 10ꢀ4
Ce0.91Nd0.09O1.955
Ce0.84Nd0.16O1.92
heating temperature
700 °C
800 °C
900 °C
CeO2
RL,0(Ce)
RL,0(Ce)
RL,0(Nd)
RL,0(Ce)
RL,0(Nd)
(2.6 ( 1.5) ꢁ 10ꢀ5
(1.1 ( 0.3) ꢁ 10ꢀ4
(3.1 ( 0.8) ꢁ 10ꢀ4
(5.9 ( 0.6) ꢁ 10ꢀ4
(1.0 ( 0.2) ꢁ 10ꢀ3
(1.8 ( 1.1) ꢁ 10ꢀ5
(6 ( 2) ꢁ 10ꢀ5
(2.7 ( 1.2) ꢁ 10ꢀ5
(5 ( 3) ꢁ 10ꢀ5
Ce0.91Nd0.09O1.955
(1.2 ( 0.1) ꢁ 10ꢀ4
(4.7 ( 0.4) ꢁ 10ꢀ4
(8.7 ( 1.0) ꢁ 10ꢀ4
(1.1 ( 0.3) ꢁ 10ꢀ4
(5.2 ( 0.4) ꢁ 10ꢀ4
(1.3 ( 0.2) ꢁ 10ꢀ3
Ce0.84Nd0.16O1.92
heating temperature
1000 °C
1100 °C
1200 °C
CeO2
RL,0(Ce)
RL,0(Ce)
RL,0(Nd)
RL,0(Ce)
RL,0(Nd)
(8 ( 6) ꢁ 10ꢀ6
(6 ( 3) ꢁ 10ꢀ6
(1.4 ( 0.3) ꢁ 10ꢀ5
(6 ( 2) ꢁ 10ꢀ5
(2.2 ( 0.4) ꢁ 10ꢀ4
ND
Ce0.91Nd0.09O1.955
(7 ( 2) ꢁ 10ꢀ5
(6 ( 2) ꢁ 10ꢀ5
(1.2 ( 0.2) ꢁ 10ꢀ4
(4.8 ( 0.3) ꢁ 10ꢀ4
(9.1 ( 1.0) ꢁ 10ꢀ4
(1.3 ( 0.2) ꢁ 10ꢀ4
(1.6 ( 0.9) ꢁ 10ꢀ4
(5.7 ( 0.6) ꢁ 10ꢀ4
Ce0.84Nd0.16O1.92
ND
ND: not determined.
9068
dx.doi.org/10.1021/ic201269c |Inorg. Chem. 2011, 50, 9059–9072

Inorganic Chemistry
ARTICLE
Figure 10. Variation of the initial normalized dissolution rate RL,0(Ce) of CeO2 (A), of Ce0.91Nd0.09O1.955 (B), and of Ce0.84Nd0.16O2 (C) and of the
long-term normalized dissolution rate RL,t(Ce) of CeO2 (D), of Ce0.91Nd0.09O1.955 (E), and of Ce0.84Nd0.16O2 (F) versus the firing temperature
(2 M HNO3, T = 60 °C).
significantly from 300 to 800 °C for the three compositions
examined. For instance, the RL,0(Ce) values reached (5.1 (
0.2) ꢁ 10ꢀ4 and (6 ( 2) ꢁ 10ꢀ5 g mꢀ2 dꢀ1 when leaching
This two-step variation of the initial normalized dissolution
rate with the heating temperature was directly connected to that
observed for the crystallite growth. Indeed, in the first tem-
perature range (300 °C < T < 800 °C), mainly associated with
the elimination of crystal defects and of amorphous domains,
the improvement of the crystallization state played an impor-
tant role on the normalized dissolution rate. Once again, such
an effect was probably linked to the decrease of the energy of
cohesion and, consequently, to the presence of such crystal
defects.
3
3
powdered Ce0.91Nd0.09O1.955 samples initially prepared at 300 and
800 °C, respectively. For CeO2 and Ce0.91Nd0.09O1.955, the initial
normalized dissolution rate was divided by a factor of 6.1 to 8.5
when increasing the firing temperature from 300 to 800 °C. This
factor was less important for Ce0.84Nd0.16O1.92, probably due to the
higher normalized dissolution rates determined. Above 800 °C, the
normalized dissolution rate values were almost stabilized.
9069
dx.doi.org/10.1021/ic201269c |Inorg. Chem. 2011, 50, 9059–9072

Inorganic Chemistry
ARTICLE
Table 6. Long-Term Normalized Dissolution Rates RL,t(Ce) and RL,t(Nd) (Expressed in g mꢀ2
Dissolution of CeO2, Ce0.91Nd0.09O1.955, and Ce0.84Nd0.16O1.92 for Several Calcination Temperatures (2M HNO3, T = 60 °C)
d
ꢀ1) Obtained during the
3
3
heating temperature
300 °C
400 °C
500 °C
CeO2
RL,t(Ce)
RL,t(Ce)
RL,t(Nd)
RL,t(Ce)
RL,t(Nd)
(1.2 ( 0.3) ꢁ 10ꢀ5
(2.5 ( 0.2) ꢁ 10ꢀ5
(1.1 ( 0.4) ꢁ 10ꢀ5
(1.9 ( 0.3) ꢁ 10ꢀ4
ND
(7 ( 3) ꢁ 10ꢀ6
(1.6 ( 0.2) ꢁ 10ꢀ5
(1.3 ( 0.1) ꢁ 10ꢀ5
(4.0 ( 0.3) ꢁ 10ꢀ5
ND
(7 ( 3) ꢁ 10ꢀ6
(1.5 ( 0.2) ꢁ 10ꢀ5
(9.0 ( 4.0) ꢁ 10ꢀ6
(1.1 ( 0.3) ꢁ 10ꢀ4
ND
Ce0.91Nd0.09O1.955
Ce0.84Nd0.16O1.92
heating temperature
600 °C
700 °C
800 °C
CeO2
RL,t(Ce)
RL,t(Ce)
RL,t(Nd)
RL,t(Ce)
RL,t(Nd)
(8 ( 3) ꢁ 10ꢀ6
(1.4 ( 0.2) ꢁ 10ꢀ5
(8.8 ( 5.0) ꢁ 10ꢀ6
(1.5 ( 0.3) ꢁ 10ꢀ4
ND
(8 ( 3) ꢁ 10ꢀ6
(1.8 ( 0.2) ꢁ 10ꢀ5
(1.7 ( 0.2) ꢁ 10ꢀ5
(1.3 ( 0.3) ꢁ 10ꢀ4
ND
(5 ( 3) ꢁ 10ꢀ6
(1.8 ( 0.2) ꢁ 10ꢀ5
(1.6 ( 0.2) ꢁ 10ꢀ5
(1.1 ( 0.3) ꢁ 10ꢀ4
ND
Ce0.91Nd0.09O1.955
Ce0.84Nd0.16O1.92
heating temperature
900 °C
1000 °C
1100 °C
CeO2
RL,0(Ce)
RL,0(Ce)
RL,0(Nd)
RL,0(Ce)
RL,0(Nd)
(8 ( 3) ꢁ 10ꢀ6
(1.3 ( 0.2) ꢁ 10ꢀ5
(6.7 ( 3.0) ꢁ 10ꢀ6
(1.4 ( 0.3) ꢁ 10ꢀ5
NDa
(3 ( 3) ꢁ 10ꢀ6
(1.7 ( 0.2) ꢁ 10ꢀ5
(3.7 ( 0.8) ꢁ 10ꢀ5
ND
(4 ( 3) ꢁ 10ꢀ6
(2.9 ( 0.2) ꢁ 10ꢀ5
(3.5 ( 0.5) ꢁ 10ꢀ5
ND
Ce0.91Nd0.09O1.955
Ce0.84Nd0.16O1.92
NDa
NDa
a ND: not determined.
average RL,t values reached 7.2 ꢁ 10ꢀ6 g mꢀ2 dꢀ1 (Ce) for
3
ꢀ3 5
CeO2, 1.6 ꢁ 10ꢀ5 g mꢀ2
d
ꢀ1 (Ce) and 1.7 ꢁ 10 g mꢀ2 dꢀ1
3
3
3
3
3
(Nd) for Ce0.91Nd0.09O1.955, and 1.3 ꢁ 10ꢀ4 g mꢀ2
d
ꢀ1 (Ce)
3
for Ce0.84Nd0.16O1.92 (Table 6, Figure 10). Conversely to what
was noted for the initial normalized dissolution rates, the firing
temperature had no effect on these long-term normalized
dissolution rates despite the elimination of crystal defects then
the increase of crystallite size. This result was mainly explained by
the presence of a neoformed amorphous gelatinous layer formed
onto the surface of the leached samples (Figure 11). As observed
on ESEM micrographs, such a phase was found to form a 20 nm
layer surrounding the initial grains. The complete characteriza-
tion of such a phase that seemed to act as a diffusion barrier
during the elementary releases is now under progress especially
by making complementary leaching tests on well-densified CeO2
and Ce1ꢀxNdxO2ꢀx/2 samples.
Figure 11. ESEM micrograph of the CeO2 powdered sample leached
for 2 months in 2 M HNO3 showing the presence of a gelatinous layer at
the surface of the crystallites (this layer was unstable under the electron
beam).
On the contrary, increasing the heating temperature above this
range (900 °C < T < 1200 °C) had no significant effect on the
chemical durability of the powders. Because the initial normal-
ized dissolution rates remained almost constant while the crystal-
lite size was significantly increased, both parameters seemed to be
independent.
Moreover, all the experiments performed using static condi-
tions exhibited two tendencies, as already described. The de-
crease from the initial normalized dissolution rates, RL,0, to the
lower values obtained near saturation conditions, RL,t, was
observed for both cerium and neodymium, for all the samples,
and all the firing temperatures. However, due to the higher
reactive surface between the solid and the solution for the lower
temperatures of calcination, the duration of the kinetics control
was significantly reduced compared to the results obtained for
samples heated at higher temperatures. The influence of the
firing temperature on the long-term normalized dissolution rates
was thus followed. These latter were significantly affected by the
chemical composition of the prepared solid solution. Indeed, the
4. CONCLUSION
Several physicochemical properties of CeO2 and CeIV
1ꢀx
NdIIIxO2ꢀx/2 mixed oxides were studied in the field of their
influence on their chemical durability. A summary of their effect
on the normalized dissolution rates is reported in Figure 12.
Some of them (temperature, acidity, chemical composition),
whose influences were usually described in the literature for other
materials, suggested that the dissolution was mainly driven by
surface reactions occurring at the solid/liquid interface. More-
over, the incorporation of trivalent lanthanide in the fluorite
structure strongly affected the chemical durability of such mixed
oxides due to the correlated formation of oxygen vacancies. The
effects of some others, such as crystal defects or crystallite size,
were never discussed in the literature for oxide samples. Although
the initial normalized dissolution rates slightly depended on the
elimination of crystal defects for firing temperatures below
800 °C, they were mainly independent of the crystallite size
9070
dx.doi.org/10.1021/ic201269c |Inorg. Chem. 2011, 50, 9059–9072

Inorganic Chemistry
ARTICLE
(5) Arab-Chapelet, B.; Grandjean, S.; Nowogrocki, G.; Abraham, F.
J. Alloys Compd. 2007, 444ꢀ445, 387–390.
(6) Matthews, R. B.; Davies, N. C. OSTI Technical Report, No.
PNL3210; Office of Scientific and Technical Information, U.S. Depart-
ment of Energy: Oak Ridge, TN, 1979.
(7) Hubert, S.; Barthelet, K.; Fourest, B.; Lagarde, G.; Dacheux, N.;
Baglan, N. J. Nucl. Mater. 2001, 297, 206–213.
(8) Mitra, N. K.; Mahapatra, S. S.; Chattopadhyay, A. K. J. Indian
Chem. Soc. 1983, 60, 499–501.
(9) Hingant, N.; Clavier, N.; Dacheux, N.; Barre, N.; Hubert, S.;
Obbade, S.; Taborda, F.; Abraham, F. J. Nucl. Mater. 2009, 385,
400–406.
(10) Hingant, N.; Clavier, N.; Dacheux, N.; Hubert, S.; Barre, N.;
Podor, R.; Aranda, L. Powder Technol. 2011, 208, 454–460.
(11) Oudinet, G.; Munoz-Viallard, I.; Aufore, L.; Gotta, M. J.;
Becker, J. M.; Chiarelli, G.; Castelli, R. J. Nucl. Mater. 2008, 375, 86–94.
(12) Kim, H. S.; Joung, C. Y.; Lee, B. H.; Oh, J. Y.; Koo, Y. H.;
Heimgartner, P. J. Nucl. Mater. 2008, 378, 98–104.
(13) Shannon, R. D. Acta Crystallogr. 1976, 32, 751–767.
(14) Lopez, C.; Deschanels, X.; Bart, J. M.; Boubals, J. M.; Den
Auwer, C.; Simoni, E. J. Nucl. Mater. 2003, 312, 76–80.
(15) Cetiner, Z. S. Environ. Monit. Assess. 2009, 151, 279–287.
(16) Dacheux, N.; Clavier, N.; Robisson, A. C.; Terra, O.; Audubert,
F.; Lartigue, J. E.; Guy, C. C. R. Chim. 2004, 7, 1141–1152.
(17) Terra, O.; Dacheux, N.; Audubert, F.; Podor, R. J. Nucl. Mater.
2006, 352, 224–232.
(18) Burghartz, M. J. Alloys Compd. 1998, 271ꢀ273, 544–548.
(19) Momin, A. C. J. Nucl. Mater. 1991, 185, 308–310.
(20) Tsuji, T. J. Alloys Compd. 1998, 271ꢀ273, 391–394.
(21) Taylor, D. Trans. J. Br. Ceram. Soc. 1984, 83, 32.
(22) Benedict, U. Physica B+C 1980, 102, 303–307.
(23) Fournier, S. Etude de la dissolution des oxydes mixtes
(U,Pu)O2 ꢀa forte teneur en plutonium. Ph.D. Thesis, No. 00 MON2
0203, Universitꢁe Montpellier II, Montpellier, France, 2000.
(24) Nikitina, G. P.; Zhukova, I. N.; Egorova, V. P. Radiochemistry
1995, 37, 193–196.
Figure 12. Influence of studied physicochemical parameters on the
CeO2 normalized dissolution rate.
(T g 900 °C). Finally, near the establishment of thermodynamic
equilibrium, the resulting normalized dissolution rates were less
affected due to the formation of a gelatinous protective layer at
the solid/liquid interface.
In order to complete this study, the effect of some other
physicochemical parameters is now under progress. For instance,
the correlations between variation of the densification rate
observed when varying the operating conditions of CeIV
1ꢀx
NdIIIxO2ꢀx/2 sintering and the chemical durability of the leached
pellets are now under study.
Comparative leaching tests of CeIV1ꢀxNdIIIxO2ꢀx/2 and
ThIV1ꢀxNdIIIxO2ꢀx/2 are also under progress to confirm that
no significant redox reaction involving the reduction of Ce(IV)
into Ce(III) occurred during these dissolution experiments.
(25) Uriarte, A. L.; Rainey, R. H. Dissolution of high-density UO2,
PuO2 and UO2-PuO2 pellets in inorganic acids; Technical Report ORNL-
3695; 1965; DOI: 10.2172/4652087.
’ AUTHOR INFORMATION
(26) Juillet, F.; Adnet, J. M.; Gasgnier, M. J. Radioanal. Nucl. Chem.
1997, 224, 137–143.
(27) Berger, P. Etude du mꢁecanisme de dissolution par oxydorꢁeduc-
tion chimique et ꢁelectrochimique des bioxydes d’actinides (UO2, NpO2,
PuO2, AmO2) en milieu aqueux acide. Ph.D. Thesis, No. 1988 PA06
6073, Universitꢁe Paris VI, Paris, 1988.
Corresponding Author
*Phone: + 33 4 66 33 92 05. Fax: + 33 4 66 79 76 11. E-mail:
nicolas.dacheux@univ-montp2.fr.
(28) Machuron-Mandard, X.; Madic, C. J. Alloys Compd. 1994,
213ꢀ214, 100–105.
’ ACKNOWLEDGMENT
(29) Joret, L.; Cote, G.; Bauer, D. Hydrometallurgy 1997, 45, 1–12.
(30) Nikitina, G. P.; Gridasov, G. C.; Listopadov, A. A.; Zhukova,
I. N.; Yakovlev, V. V. Radiochemistry 1995, 37, 365–371.
(31) Lide, D. R. Handbook of Chemistry and Physics, 71st ed.; CRC
Press: Boca Raton, FL, 1990ꢀ1991.
The authors would like to thank Adel Mesbah for his help in
the refinement of XRD data. They also thank the MATINEX
French Research Group (Innovative materials in extreme con-
ditions, CEA/CNRS/AREVA/EDF/French Universities) in-
cluded in the PACEN Program for their subsequent financial
support. This work also benefited from the financial support of
the French National Research Agency (ANR, project # ANR-08-
BLAN-0216) and from the CNRS Interdisciplinary Research
Program MaProSu (Matꢁeriaux et Procꢁedꢁes de Remplacement/
Substitution).
(32) Joseph, C.; Varghese, G.; Ittyachen, A. J. Therm. Anal. Calorim.
1998, 52, 517–522.
(33) Ollendorff, N.; Weigel, F. Inorg. Nucl. Chem. Lett. 1969, 5,
263–269.
(34) Finger, L. W. J. Appl. Crystallogr. 1998, 31, 111.
(35) Roisnel, T.; Rodriguez-Carvajal, J. Mater. Sci. Forum 2001,
378ꢀ381, 118–123.
(36) Barret, E. P.; Joyner, L. G.; Halenda, P. H. J. Am. Chem. Soc.
1951, 73, 373–380.
’ REFERENCES
(1) Ichimiya, M.; Sagayama, Y. Trans. Am. Nucl. Soc. 2004,
90, 46–47.
(2) Hoffelner, W. Chimia 2005, 59, 977–982.
(3) Tommasi, J.; Delpech, M.; Groullier, J. P.; Zaetta, A. Nucl.
Technol. 1995, 111, 133–148.
(37) Dacheux, N.; Le Du, J. F.; Brandel, V.; Genet, M.; Decambox,
P.; Moulin, C. New J. Chem. 1996, 20, 507–514.
(38) Thomas, A. C.; Dacheux, N.; Le Coustumer, P.; Brandel, V.;
Genet, M. J. Nucl. Mater. 2000, 281, 91–105.
(39) Lasaga, A. C. J. Geophys. Res. 1989, 89, 4009–4025.
(40) Dacheux, N.; du Fou de Kerdaniel, E.; Clavier, N.; Podor, R.;
Aupiais, J.; Szenknect, S. J. Nucl. Mater. 2010, 404, 33–43.
(4) Buisson, P. Ph.D. Thesis, GRE1-0519, University Grenoble 1,
Grenoble, France, 1999.
9071
dx.doi.org/10.1021/ic201269c |Inorg. Chem. 2011, 50, 9059–9072

Inorganic Chemistry
ARTICLE
(41) Dacheux, N.; Clavier, N.; Ritt, J. J. Nucl. Mater. 2006,
349, 291–303.
(80) Vandenborre, J.; Grambow, B.; Abdelouas, A. Inorg. Chem.
2010, 49, 8736–8748.
(42) Clavier, N.; du Fou de Kerdaniel, E.; Dacheux, N.; Le Coustu-
mer, P.; Drot, R.; Ravaux, J.; Simoni, E. J. Nucl. Mater. 2006, 349,
304–316.
(43) Walther, J. V. Am. J. Sci. 1996, 296, 693–728.
(44) Ganor, J.; Mogollon, J. L.; Lasaga, A. C. Geochim. Cosmochim.
Acta 1995, 59, 1037–1052.
(45) Pokrovsky, O. S.; Schott, J. Geochim. Cosmochim. Acta 2000,
64, 3313–3325.
(46) Furrer, G.; Stumm, W. Geochim. Cosmochim. Acta 1991,
55, 2193–2201.
(47) Heisbourg, G.; Hubert, S.; Dacheux, N.; Ritt, J. J. Nucl. Mater.
2003, 321, 141–151.
(48) Heisbourg, G.; Hubert, S.; Dacheux, N.; Purans, J. J. Nucl.
Mater. 2004, 335, 5–13.
(49) Chou, L.; Wollast, R. The Chemistry of Weathering; Drever, J. I.,
Ed.; 1985.
(50) Chou, L.; Wollast, R. Am. J. Sci. 1985, 258, 963–993.
(51) Furrer, G.; Stumm, W. Geochim. Cosmochim. Acta 1986,
50, 1847–1860.
(52) Blum, A. E.; Lasaga, A. C. Nature 1988, 331, 431–433.
(53) Robisson, A. C.; Dacheux, N.; Aupiais, J. J. Nucl. Mater. 2002,
306, 134–146.
(54) Grenthe, I.; Puigdomenech, I. Modelling in Aquatic Chemistry;
OECD Publications: Paris, 1997.
(55) Eyring, H. J. Chem. Phys. 1935, 3, 107–115.
(56) Rose, N. M. Geochim. Cosmochim. Acta 1991, 55, 3273–3286.
(57) Bruno, J.; Casas, I.; Puigdomꢀenech, I. Geochim. Cosmochim.
Acta 1990, 55, 647–658.
(58) De Pablo, J.; Casas, I.; Gimenez, J.; Molera, M.; Rovira, M.;
Duro, L.; Bruno, J. Geochim. Cosmochim. Acta 1999, 63, 3097–3103.
(59) Hubert, S.; Heisbourg, G.; Dacheux, N.; Moisy, P. Inorg. Chem.
2008, 47, 2064–2073.
(60) Ubaldini, A.; Artini, C.; Costa, G. A.; Carnasciali, M. M.; Masini,
R. J. Therm. Anal. Calorim. 2006, 84, 207–211.
(61) Higashi, K.; Sonoda, K.; Ono, H.; Sameshima, S.; Hirata, Y.
J. Mater. Res. 1999, 14, 957–967.
(62) Horlait, D.; Claparꢀede, L.; Clavier, N.; Szenknect, S.; Dacheux,
N.; Ravaux, J.; Podor, R. Inorg. Chem. 2011, 50, 7150–7161.
(63) Sharov, V. A.; Bezdenezhnykh, G. V. Usp. Khim. 1981, 50, 1197.
(64) Balboul, B. A. A.; Myhoub, A. Y. Z. J. Anal. Appl. Pyrolysis 2010,
89, 95–101.
(65) Hong, S. J.; Virkar, A. V. J. Am. Ceram. Soc. 1995, 78, 433–439.
(66) JCPDS Card No. 010810792.
(67) Ikuma, Y.; Shimada, E.; Okamura, N. J. Am. Ceram. Soc. 2005,
88, 419–423.
(68) Mikulova, J.; Rossignol, S.; Gꢁerard, F.; Mesnard, D.; Kappen-
stein, C.; Duprez, D. J. Solid State Chem. 2006, 179, 2511–2520.
(69) Guo, X.; Waser, R. Prog. Mater. Sci. 2006, 51, 151–210.
(70) Ostwald, W. Z. Phys. Chem. 1897, 22, 289–302.
(71) Le Coustumer, P.; Monthioux, M.; Oberlin, A. J. Eur. Ceram.
Soc. 1993, 11, 95–103.
(72) Belliꢀere, V.; Joorst, G.; Stephan, O.; de Groot, F. M. F.;
Weckhuysen, B. M. J. Phys. Chem. B 2006, 110, 9984–9990.
(73) Jarcho, M.; Bolen, C. H.; Thomas, M. B.; Bobick, J.; Kay, J. F.;
Doremus, R. H. J. Mater. Sci. 1976, 11, 2027–2035.
(74) Chen, J. C.; Chen, W. C.; Tien, Y. C.; Shih, C. J. J. Alloys Compd.
2010, 496, 364–369.
(75) Dollimore, D. Thermochim. Acta 1987, 117, 331–363.
(76) Wintergerst, M.; Dacheux, N.; Datcharry, F.; Herms, E.;
Kapusta, B. J. Nucl. Mater. 2009, 393, 369–380.
(77) Ruas, A.; Pochon, P.; Simonin, J. P.; Moisy, P. Dalton Trans.
2010, 39, 10148–10153.
(78) Krawetz, A. Raman spectral study of equilibria in aqueous
solutions of nitric acid. Ph.D. Thesis, University of Chicago, Chicago, IL,
1955.
(79) Terry, B. Hydrometallurgy 1983, 11, 315–344.
9072
dx.doi.org/10.1021/ic201269c |Inorg. Chem. 2011, 50, 9059–9072